Researchers and counselors suggest that effects of psychological abuse may be more debilitating and longlasting to women than physical abuse. While psychological abuse may precede and accompany physical violence, empirical evidence is virtually nonexistent. As presently defined, abuse occurs when a male partner expresses behaviors, attitudes and beliefs that damage a female's self-concept, restrict her self-determination, keep her in a harmful relationship or decrease her ability to function. The theoretical under-pinnings suggest that abuse can occur during ordinary interaction between partners. Effects and processes may escalate, becoming more overtly abusive and damaging. Specifically, the research proposes to describe psychological abuse empirically, distinguishing and defining three levels: violation (common), harassment (more serious, less common) and battering (systematic, severe). Surveys will be sent to counselors (N = 600) and physical abuse specialists to obtain their views on the three levels of abuse referring to those in functional, dysfunctional and physically abusive relationships. The responses will be incorporated into an interview protocol. Participants will be recruited through, e.g., newspapers. Women (N = 160) self- identified as psychologically or physically abused will be paid to complete several questionnaires and an in-depth interview. Questions will probe perceptions, feelings and descriptions of common and uncommon events in their partnerships which made them feel good and bad about themselves, their relationship and partner. Transitory and longlasting effects will be examined as will ways in which her partner expresses jealousy, anger, love, etc. Other measures (e.g., self-concept) will be included. This phase allows assessment of differences between psychological and physical abuse. Phases 3 and 4 are longitudinal to assess possible tansitory aspects or escalation effects. Two interviews a year apart will be conducted with women (N = 300) screened so that equal numbers will represent three levels of marital adjustment with no evidence of physical abuse. Two groups will be composed of self-identified psychologically and physically abused women. Including five functional and dysfunctional groups allows appropriate comparisons for each proposed level of psychological abuse. Finally, observations will be conducted using student couples to test hypotheses derived during the project. These data allow comparisons between forms of abuse in different types of relationships and within groups change over time. The research will generate operational definitions of violation, harassment and battering and a model of psychological abuse.